[computer-go] analysis of UCT and BAST

I have updated the BAST paper, providing additional comparison with UCT, as suggested by one person in the list. See: https://hal.inria.fr/inria-00150207 Basically, in the considered examples, BAST with an appropriate smoothness sequence performs always better than both UCT and Flat-UCB. Now,

Re: [computer-go] analysis of UCT and other bandit algorithms for tree search

of the result, but rather helping people decide how interested they'd be in the paper. There are likely counters to the points above. For example, the addresses #1 by saying the minimax problem is a direct generalization of the results presented On 4/24/07, Remi Munos [EMAIL PROTECTED] wrote

Re: [computer-go] analysis of UCT and other bandit algorithms for tree search

Le Mercredi 25 Avril 2007 15:49, vous avez écrit : On 4/25/07, Remi Munos [EMAIL PROTECTED] wrote: 3. The goal is to find the optimal leaf rather than the optimal subtree. I'm not sure to understand this comment. Actually the goal is to find a close-to optimal play at the root level

[computer-go] analysis of UCT and other bandit algorithms for tree search

-max tree case but extension to min-max search is straightforward. Best, Remi Munos ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/